Power Dissipation Index (PDI)
Power Dissipation Index (PDI) for a tropical cyclone is defined
as, "the sum of the maximum one-minute sustained wind speed
cubed, at six-hourly intervals, for all periods when the cyclone
is at least tropical storm strength". I apply a scaling factor so
as to match the scale that Emanuel uses.
Data on each storm's "maximum one-minute sustained
windspeed" is available from the United States National
Hurricane Center at their website. This data is given in sixhour intervals. This is the data used.
Example of PDI Calculation:
Choose Tropical Storm Arlene, which was the first storm of
the 2005 season.
The storm had:
3 six-hour periods with a max windspeed of 40 mph
1 six-hour period with a max windspeed at 45 mph
1 six-hour period with a max windspeed at 50 mph
2 six-hour periods with a max windspeed of 60 mph
2 six-hour periods with a max windspeed of 65 mph
2 six-hour periods with a max windspeed of 70 mph
And that's it. All other periods were 35 mph or less and are not
counted.
For the 40 mph period, calculate the cube of 40. Then, multiply
that cube by 3.
(40x40x40) (3) = 192,000
Do the same for the other windspeeds, and then add the results
Unscaled PDI for Arlene = 192,000 + 91,125 + 125,000 +
432,000 + 549,250 + 686,000
= 2,075,375
 To get the scaling factor, calculate the unscaled PDI for a
year for which Emanuel's PDI is known. Then calculate
the scaling factor. Then, confirm that this scaling factor is
correct by calculating the unscaled PDI for two other
years and multiplying each by the calculated scaling
factor, and confirming that the result reasonably matches
Emanuel's PDIs.
Calculated unscaled PDI for the storms of 1997 = 46,150,000
Emanuel's PDI for 1997 = 4.15
Scaling factor = 4.15 divided by 46.15 million = 8.99 x 10 to
minus 8
Check: Unscaled PDI calculated for the storms of 1993 =
37,930,000
Multiply this unscaled PDI by the scaling factor giving a
scaled PDI result of 3.37
Compare with Emanuel's PDI of 3.50
Slightly less than -4% difference: close, but check again
Check: Unscaled PDI calculated for the storms of 2000 =
139,020,000
Multiply this unscaled PDI by the scaling factor, giving a
scaled PDI result of 12.50
Compare with Emanuel's PDI of 12.20
About +2.5% difference. Close enough for our purposes.
 Examine the 2005 season, the season which had a
record number of storms. The unscaled PDI for
2005 is 338,180,000. I don't show the calculations
here because they are massive, but here are the
numbers in each wind category:
40 mph: 75
45 mph: 51
50 mph: 84
55 mph: 0
60 mph: 75
65 mph: 55
70 mph: 49
75 mph: 49
80 mph: 21
85 mph: 26
90 mph: 13
95 mph: 0
100 mph: 10
105 mph: 10
110 mph: 12
115 mph: 12
120 mph: 2
125 mph: 14
130 mph: 0
135 mph: 5
140 mph: 9
145 mph: 6
150 mph: 9
155 mph: 3
160 mph: 4
165 mph: 1
170 mph: 0
175 mph: 3
180 mph: 0
185 mph: 1
 For the 2006 season to-date (October 6), the
unscaled PDI is 80,650,000, which corresponds to a
scaled PDI of 7.25 (note: the windspeeds are
unofficial numbers and are subject to reanalysis
after the season ends.)
 The season has not ended, so estimate that there
will be one more tropical storm (unscaled PDI =
2,000,000) and one weak hurricane (unscaled PDI =
5,000,000) in 2006, These two storms combined
have an unscaled PDI of 7,000,000, which equates
to a scaled total of 0.63. Add this to 7.25 and the
total estimated PDI for 2006 is 7.25 + 0.65 = 7.90